Compare / Kimi K2 Thinking vs Nemotron 3 Nano 30B A3B
Kimi K2 Thinking vs Nemotron 3 Nano 30B A3B
Which is cheaper, which is more capable, and which does better at each kind of work, from 16 results on 3 sources that measured both models.
Kimi K2 Thinking or Nemotron 3 Nano 30B A3B? The short answer
- Nemotron 3 Nano 30B A3B is cheaper: $0.088 against $1.07 per million tokens, 12 times less.
- Kimi K2 Thinking scores higher on the Artificial Analysis Intelligence Index: 22.0 against 8.9.
- On our business benchmarks, Kimi K2 Thinking ranks higher for agents and tool use, reasoning and knowledge and writing.
At a glance
| Kimi K2 Thinking | Nemotron 3 Nano 30B A3B | |
|---|---|---|
| Price per million tokens | $0.60 in · $2.50 out | $0.050 in · $0.20 out |
| Blended (3 in : 1 out) | $1.07 | $0.088 |
| Cheapest host, blended | $1.07 (Google) | $0.088 (Crusoe, FP8) |
| Intelligence Index | 22.0 | 8.9 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | 220 |
| Context window | 262,144 tokens | 262,144 tokens |
| Weights | Open | Open |
| Developer based in | China | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Kimi K2 Thinking | Nemotron 3 Nano 30B A3B |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 133 of 161 | 145 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 82 of 195 | 130 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 48 of 187 | 169 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | not measured | 121 of 153 |
Every shared result
| Benchmark | Metric | Kimi K2 Thinking | Nemotron 3 Nano 30B A3B |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 22.0 | 8.90 |
| Artificial Analysisreported | GPQA Diamond% of questions | 83.8% | 75.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 23.8% | 11.4% |
| Artificial Analysisreported | IFBench% of instructions | 68.1% | 71.1% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 72.0% | 38.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 31.1% | 13.6% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 93.0% | 40.9% |
| OpenHands Indexreported | Average% resolved | 41.0% | 15.5% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 32.4% | 20.9% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 12.5% | 6.2% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 43.6% | 8.5% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 69.2% | 34.2% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 47.3% | 7.6% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 20.0% | 35.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.38 | 0.28 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 59.3 | 25.9 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Kimi K2 Thinking better than Nemotron 3 Nano 30B A3B?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Kimi K2 Thinking ranks higher for agents and tool use, reasoning and knowledge and writing.
Which is cheaper, Kimi K2 Thinking or Nemotron 3 Nano 30B A3B?
Nemotron 3 Nano 30B A3B costs $0.088 per million tokens (three input to one output), against $1.07 for Kimi K2 Thinking.
Run Kimi K2 Thinking and Nemotron 3 Nano 30B A3B on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.